ANEXO AL TEMA 4 FÍSICA DEL ESTADO SÓLIDO II

BANDSTRUCTURE: SILICON
BANDSTRUCTURE: GaAs
Although the bandstructure of Si is far from ideal, having an indicrect bandgap, hig hhole masses,
and small spin-orbit splitting, processing related advantages
The bandgapmakeat 0 KSiisthe
1.51premier
eV and semiconductor
ESTRUCTURA
for consumer electronics.DEOn theBANDAS
right we show E k =INTERVALO
constant energy
(0,0,0), G-point.DE
ellipsoids
i.e., the for ENERGÍA
TheSiupper
conduction band.
conduction PROHIBIDA
band valleys are at the B PARA ALGUNOS
at 300 K it is 1.43 eV. The bottom of the conductionband is at
BANDSTRUCTURE
ANDSTRUCTURE
L-point. : Ge AlAs, SEMICONDUCTORES
InAs,InAs,
: Ge AlAs, InP InP
There are six equivalent valleys in Si at the bandedge.
Germanium
Germanium Aluminum ArsenideArsenide
Aluminum
Silicon Gallium Arsenide 6 6 6 6
6 Eg = 0.66eVEg = 0.66eV Eg = 2.15eVEg = 2.15eV
6 5 5 5 5at 300K at 300K
BANDSTRUCTURE5: SILICON Eg = 1.1eV
Eg = 1.43eV
at 300K at 300K
at 300K CONDUCTION 4 4 4 4
5 at 300 K BAND
4 an indicrect bandgap, hig hhole masses,
re of Si is far from ideal, having kz 3 3 3 3
ting, processing related advantages make Si the premier semiconductor
On the right we show constant energy ellipsoids for Si conduction band.
BANDSTRUCTURE
4 : GaAs 2 2 2 2
(001)

ENERGY

ENERGY
3

ENERGY

ENERGY
valleys in Si at the bandedge. 3 0.3 0.3
ENERGY (eV)

2.751 Eg2.75

(eV)

(eV)
1 1 Eg

(eV)

(eV)
1

ENERGY (eV)
2 The3.4 bandgap at 0 K is 1.51 eV and at 300 K it is 1.43 2 eV. The bottomSix of the conductionband
equivalent is at Eg Eg
0.90 0.90
1.1 k = (0,0,0), i.e., the G-point. The upper conduction band 0.3 valleys are 0.58
at the L-point. 0 0 0 0
(010) valleys at
1 1 (100) 5.4 x 10 –4 T 2
V
Eg Eg conduction Eg = 1.519 – –1 –1 (eV) –1 –1
bandedge T + 204
0kz 0 –2 –2 –2 –2
Gallium Arsenide T = Temperature in K
–3 –3 –3 –3
–1 (001) kx (100) –1 (010) k
6 y
–4 –4
–4 –4
Six equivalent Eg = 1.43eV
CONDUCTION (001)
–2 (a) (a) L [111] ΓL [100] [111]XΓ [100]X
(b) (b) L [111] Γ L [100]
[111]XΓ [100]X
–2 5
valleys at at 300 K
(010) (100) BAND
conduction –3
–3 4
bandedge 4.37 x 10–4 T2 Indium Arsenide
Indium Arsenide Indium Phosphide
Indium Phosphide
Eg = 1.17 – –4 (eV) T = Temperature in K 6 6
kx (100) –4 3 T – 636 [111] [100] 6 6
(010) ky Eg = 0.35eVEg = 0.35eV Eg = 1.34eVEg = 1.34eV
ENERGY (eV)

(001)
k k 5 5300K 5 5
2 at at 300K at 300K at 300K
0.3 0.58 (b)
(a) 4 4 4 4
4.37 x 10–4 T2 1 10–4 T2mass is light.
5.4• xElectron
Eg = 1.17 –
T – 636
(eV) T = Temperature in KEg C ONDUCTION BAND
Eg = 1.519 – : (eV) m* = 0.067
3 m 0 3 3 3
T + 204 valley mass is large. m* = 0.25 m
weak optical transitions, cannot be used to• Upper results in negative

ENERGY

ENERGY
ENERGY

ENERGY
• Indirect gap material 0 produce lasers. 2 0 2 2 2 0.6 0.9

(eV)
0.6

(eV)
T = Temperaturedifferential
in K 0.9

(eV)
resistance at higher fields.

(eV)
(b)
9 9 4 4
• Valleys along the x-axis
2π
–1 and –x-axis: k = (0.85,0,0) and k = (–0.85,0,0): 2π • Material is direct bandgap and has strong
1 optical1 transistions can 1 1 Eg Eg
0x a 0x a be used for light emission. E E
weak optical transitions, cannot be used to produce –2lasers. 0 0g g 0 0
h 2 ( k x – k 0x ) 2 k 2 + k2
x z
s and –x-axis: k0x = a (0.85,0,0) and k0xE(k)
2π 2π = Ec +
= a (–0.85,0,0): + * VALENCE ; ml =BAND : m0; mt• =Heavy
0.98 0.19hole
m0 mass: 0.45 m0; light hole–1mass = –1 0.08 m0. –1 –1
–32 m*l mt Intrinsic carrier concentration at 300 = 1.84 x 10 6 cm–3.
h2 (kx – k0x)2 + kx + kz ; m similar
2 2 –2 –2 –2 –2
l = 0.98 mE-k
0; mtrelationsm0 for other 4 valleys.
= 0.19 –4
2 m*l m*t [111] [100] –3 –3 –3 –3
s for other 4 valleys.
• Density of states mass = 1.08 m0 k(6 valleys included). –4 –4 –4 –4
= 1.08 m0 (6 valleys included). L [111] Γ L [100]
[111]XΓ [100]X L [111] Γ L [100]
[111]XΓ [100]X
(c) (c) (d) (d) k
• Heavy hole mass: C 0.49 mBAND
ONDUCTION 0 ; light
: hole
• mass: mass
Electron 0.16 ismlight.
0. m* = 0.067 m k k k
m0 ; light hole mass: 0.16 m0. 0
• Upper valley mass is large. m* = 0.25 m0 results in negative
1010 cm–3
K: 1.5 xresistance (a) Bandstructure of Ge. (b)ofBandstructure
(a) Bandstructure of AlAs. of
Ge. (b) Bandstructure (c)AlAs.
Bandstructure of InAs. of
(c) Bandstructure Since noSince
InAs. adequate
no adequate
ration at 300 K: 1.5 x 10•10Intrinsic
cm–3. carrier concentration at 300 differential at.higher fields.
© Prof. Jasprit substitutesubstitute
matches InAs
matches directly, it is oftenitused
InAs directly, as an
is often alloy
used as (InGaAs, InAlAs, etc.,)
an alloy (InGaAs, for etc.,)
InAlAs, devices.
for devices.
• Material is direct bandgap andSingh www.eecs.umich.edu/~singh
has strong optical transistions can
(d) Bandstructure of InP. InP is a very important material material
for high speed devices
be used for light emission. (d) Bandstructure of InP. InP is a very important for high speedas well asasawell as a
devices
substratesubstrate
and barrierand layer material
barrier layer for semiconductor
material for lasers. lasers.
semiconductor
ww.eecs.umich.edu/~singh
© Prof. Jasprit SinghBAND
VALENCE www.eecs.umich.edu/~singh
: • Heavy hole mass: 0.45 m0; light hole mass = 0.08 m0.
© Prof. Jasprit
© Prof.Singh www.eecs.umich.edu/~singh
Jasprit Singh www.eecs.umich.edu/~singh
Intrinsic carrier concentration at 300 = 1.84 x 106 cm–3.
 SPLIT-OFF BAND: 2 2
Eso = –∆ – h k
2m*so
1 2
2pcv
= –1 +
m*so m0 3m02(EgΓ + ∆)

HEAVYMASA EFECTIVA
HOLE ; LIGHT HOLE : EN BANDA DE CONDUCCIÓN PARA SEMICONDUCTORES MÁS TÍPICOS
In a simple approximation the heavy hole and light hole bands can also be
represented by masses m*hh and m* h. However, the real picture is more complex.

CdS
0.2 AlAs
AlSb
( mmoc )

ZnTe GaP
*

ZnSe
CdSe
EFFECTIVE MASS

CdTe
0.1 InP
GaAs
GaSb
InAs
I Banda de huecos ligeros
EFFECTIVE MASS DESCRIPTION
InSb II Banda huecos pesados
0
0 1.0 2.0 3.0 III Split-Off banda de Valencia
BANDGAP Eg (eV)
𝛥 Split-Off energía
CONDUCTION BAND: Direct bandgap material
© Prof. Jasprit Singh www.eecs.umich.edu/~singh
2 2
Ec(k) = Ec(0) + h k
2m*c
with 2 1 2
1
m*c
= 1
m0
+
2pcv
m20 3 EgΓ
2 + 1
EgΓ + ∆ ( ) 2pcv
; m ~22 eV
0
The smaller the bandgap, the smaller the effective mass.

SPLIT-OFF BAND: 2 2
Eso = –∆ – h k
2m*so
1 2
2pcv
= –1 +
m*so m0 3m2(E + ∆)
 ELECTRONIC PROPERTIES OF SOME SEMICONDUCTORS

Material Bandgap Relative Material Electron Hole

(eV) Dielectric Mass Mass
Constant (m0) (m0)

C 5.5, I 5.57 AlAs 0.1

Si 1.124, I 11.9 AlSb 0.12 * = 0.98
mdos
Ge 0.664, I 16.2 GaN 0.19 * = 0.60
mdos
SiC 2.416, I 9.72 * = 0.60
GaP 0.82 mdos
GaAs 1.424, D 13.18
GaAs 0.067 m*lh = 0.082
AlAs 2.153, I 10.06 m*hh = 0.45

InAs 0.354, D 15.15 GaSb 0.042 * = 0.40

mdos

GaP 2.272, I 11.11 Ge ml = 1.64 * = 0.044

mlh
mt = 0.082 m*hh = 0.28
InP 1.344, D 12.56
mdos= 0.56
InSb 0.230, D 16.8
InP 0.073 * = 0.64
mdos
CdTe 1.475, D 10.2

AlN 6.2, D 9.14

InAs 0.027 * = 0.4
mdos
Propiedades de algunos semiconductores. D e I indican
InSb 0.13 m*dos = 0.4 transición directa e indirecta respectivamente. Los datos
GaN 3.44, D 10.0
Si ml = 0.98 m*lh = 0.16
mostrados corresponden con una temperatura de 300K.
ZnSe 2.822, D 9.1 * = 0.49
mt = 0.19 mhh
ZnTe 2.394, D 8.7 mdos= 1.08

Properties of some semiconductors. D and I stand for direct and indirect gap,
respectively. The data are at 300 K. Note that Si has six conducton band
valleys, while Ge has four.

© Prof. Jasprit Singh www.eecs.umich.edu/~singh

ELECTRONES Y HUECOS EN UN SEMICONDUCTOR INTRÍNSECO
108 CHAPTER 4 The Semiconductor in Equilibrium

108 CHAPTER 4 The Semiconductor in Equilibrium

E
108 CHAPTER 4 The Semiconductor in Equilibrium

E
E

E
Densidad de estados en banda
gc(E)
gc(E)fF (E) ! n(E) fF (E)
de valencia y conducción y
E
gc(E)
gc(E)fF (E) ! n(E)
gc(E)f
fF (E) E
gc(E)
función de Fermi. Las áreas
AreaF (E)
! n!!n(E) fF (E)
Ec gc(E)
electron
0
gc(E)
gc(E)
representan la concentración
concentration Ec
Ec
Ec fF (E)
Area ! n0 !
electron Area ! n0 !
0
de electrones y huecos en el
EF
concentration electron E
c
concentration Ec (b) caso de que el nivel de Fermi
fF (E)
fF (E)
0
(b)
0 esté en el centro del intervalo
EF
EF
(b)
de energía prohibida.
gv(E)(1 " fF (E)) ! p(E) Ev
Ev [1 " fF (E)]
Mostramos ampliación de las
gv(E)(1 " fF (E)) ! p(E)
gv(E)
Ev
gv(E)(1 " fF (E)) ! p(E) Ev
gv(E)
zonas cerca de los bordes de
Ev [1 " fF (E)]
Ev gv(E) E
gv(E)
[1 " fF (E)] banda de valencia y banda de
gv(E)
gv(E) Area ! p0 !
E
hole concentration conducción.
E
Area ! p0 !
hole concentration Area ! p0 !
hole concentration
fF (E) ! 0 fF (E) ! 1
(a) (c)
fF (E) ! 0 fF (E) ! 1
Figure
fF (E) ! 0 4.1 | (a) Density of states functions,
fF (E) !1 Fermi–Dirac probability function, and areas representing electron and hole
(a) for the case when E is near the midgap energy; (b) expanded
concentrations (c) view near (c)
the conduction-band energy;
F
(a)
and (c)
Figure 4.1 | (a) Density of expanded view near
states functions, the valence-band
Fermi–Dirac energy.
probability function, and areas representing electron and hole
Figure
concentrations for 4.1when
the case | (a) Density of the
EF is near states functions,
midgap Fermi–Dirac
energy; probability
(b) expanded view nearfunction, and areas representing
the conduction-band energy; electron and hole
and (c) expandedconcentrations for the case when
view near the valence-band EF is near the midgap energy; (b) expanded view near the conduction-band energy;
energy.
and (c) expanded view near the valence-band energy.
semiconductor, then, electrons and holes are created in pairs by the thermal energy so
that the number of electrons in the conduction band is equal to the number of holes
semiconductor, then, electrons
in the valenceand holes are created in pairs by the thermal energy so
band.
that the numbersemiconductor,
of electrons inthen,
Figure 4.1a
electrons and
theshows
conduction band
a plot of
holesequal
the is
are created
density of the in
to states pairs by
number
functionofthe thermal energy so
holes
in the conduction-band
that the number of electrons in the conduction band is equal to the number of holes
in the valence band.
gc(E), the density of states function in the valence-band gv(E), and the Fermi–Dirac
Figure 4.1ainshows
the valence
probability
band.
a plot function
of the density
for T of
! states function
0 K when EF isinapproximately
the conduction-band
halfway between Ec and
gc(E), the densityEof Figure 4.1a shows atheplot of the density of
(E),states
and function in the conduction-band
v. states
If we function
assume, in for thevalence-band
moment, thatgvthe the and
electron Fermi–Dirac
hole effective masses are
gc(E),for
probability function theTdensity
! 0gcK of states function in the valence-band gv(E),Eand the Fermi–Dirac
equal, then (E)when is approximately
and gEvF(E) are symmetrical halfway
functionsbetween
about the c and
midgap energy (the
probability
Ev. If we assume,energy function
for themidway
moment, for T !
that the 0 K when EF is approximately halfway between Ec and
between Ec electron
and Ev). and hole effective
We noted masses
previously that thearefunction fF (E ) for
Ev.and
equal, then gc(E) If we assume, for the moment, that the electron and hole effective masses are
E ! gEvF(E) are symmetrical
is symmetrical to thefunctions
functionabout
1 " fFthe
(E)midgap
for E #energy
EF about (the
the energy E $ EF.
equal,
energy midway between then g c(E) and gv(E) are symmetrical functions about the midgap energy (the
and Ev).that
Wethenoted previously
This also Ec means function fF(E) that
for E the$function
EF % dE fF (E
is )equal
for to the function
ELECTRONES Y HUECOS EN UN SEMICONDUCTOR INTRÍNSECO

Aproximación de Maxwell-Boltzman:
3.5 Statistical Mechanics FERMI - DIRAC 97

1
𝑓# 𝐸 =
𝐸 − 𝐸#
Fermi–Dirac function 1 + exp
𝑘𝑇
1.0

Boltzmann approximation
1
2

−(𝐸 − 𝐸# )
𝑓/0 𝐸 ≈ exp
𝑘𝑇
E EF

Figure 3.35 | The Fermi–Dirac probability function and the

MAXWELL - BOLTZMANN
Maxwell–Boltzmann approximation.

Objective: Determine the energy at which the Boltzmann approximation may be considered EXAMPLE 3.8
valid.
Calculate the energy, in terms of kT and EF, at which the difference between the
Boltzmann approximation and the Fermi–Dirac function is 5 percent of the Fermi function.

■ Solution
We can write
ELECTRONES Y HUECOS EN UN SEMICONDUCTOR INTRÍNSECO

En un semiconductor intrínseco y bajo la aproximación de Maxwell-Boltzman:

C
9 9
2𝜋𝑚?∗ 𝑘𝑇 B −(𝐸D − 𝐸# ) −(𝐸D − 𝐸# )
𝑛 = 5 𝑔7 𝐸 𝑓# 𝐸 𝑑𝐸 ≈ 5 𝑔7 𝐸 𝑓/0 𝐸 𝑑𝐸 ≈ 2 exp = 𝑁7 exp
:; :; ℎB 𝑘𝑇 𝑘𝑇
C
:H :H ∗
2𝜋𝑚J 𝑘𝑇 B −(𝐸# − 𝐸G )
𝑝 = 5 𝑔G 𝐸 (1 − 𝑓# 𝐸 )𝑑𝐸 ≈ 5 𝑔7 𝐸 (1 − 𝑓/0 𝐸 )𝑑𝐸 ≈ 2 exp =
I9 I9 ℎB 𝑘𝑇

−(𝐸# − 𝐸G )
𝑝 = 𝑁G exp
𝑘𝑇

En todo semiconductor intrínseco n =p

ELECTRONES Y HUECOS EN UN SEMICONDUCTOR INTRÍNSECO

Consecuencias:

−(𝐸D − 𝐸# ) −(𝐸# − 𝐸G )
𝑛 = 𝑝 ⇒ 𝑁7 exp = 𝑁G exp ⇒ 𝑜𝑝𝑒𝑟𝑎𝑛𝑑𝑜
𝑘𝑇 𝑘𝑇

1 1 𝑁G 1 3 𝑚S∗ 1 1
𝐸# = 𝐸G + 𝐸D + 𝑘𝑇 𝐿𝑛 = 𝐸G + 𝐸D + 𝑘𝑇 𝐿𝑛 ≈ 𝐸G + 𝐸D = 𝐸
2 2 𝑁D 2 4 𝑚?∗ 2 2 TUJ

−𝐸TUJ
𝑛 · 𝑝 = 𝑛WB = 𝑁D 𝑁G exp = 𝑐𝑡𝑒.
𝑘𝑇 Material ni (cm-3)
C/a Si 1,5 x1010
𝑚?∗ 𝑚S∗ −𝐸TUJ
𝑛W = 𝑝W = 4.826 𝑥10B_ B 𝑇 C/B exp 𝑚IC GaAs 1,8 x106
𝑚? 2𝑘𝑇
Ge 2,4 x1010
 HAPTER 4
ELECTRONES Y HUECOS EN UN SEMICONDUCTOR INTRÍNSECO
The Semiconductor in Equilibrium

1500 T(•C)
1000 500 200 100 27 0—20
1019

1018
𝜎 total = ne𝜇? + 𝑝𝑒𝜇S =
1017
C/a
Ge
B_
𝑚?∗ 𝑚S∗ −𝐸TUJ
1016
= 4.826 𝑥10 𝑇 C/B 𝑒 𝜇? + 𝜇S exp =
1015
𝑚?B 2𝑘𝑇
Intrinsic carrier density ni (cm!3)

Si
1014
C/a
𝑚?∗ 𝑚S∗ −𝐸TUJ
1013
= 773.1 𝑇 C/B 𝜇? + 𝜇S exp
1012
𝑚?B 2𝑘𝑇
1011
−𝐸TUJ
1010 𝜎 total = 𝜎j exp
GaAs 2𝑘𝑇
109

108

107

106
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
1000#T(K !1)

Concentración de portadores intrínsecos

Figure 4.2 | The intrinsic carrier
concentration of Ge, Si, and GaAs as a
en diferentes
function of semiconductores
temperature. en función
(From Sze [14].)
de la temperatura.

.1.4 The Intrinsic Fermi-Level Position

We have qualitatively argued that the Fermi energy level is located near the center
f the forbidden bandgap for the intrinsic semiconductor. We can specifically calcu-
ate the intrinsic Fermi-level position. Since the electron and hole concentrations are
 ELECTRONES Y HUECOS EN UN SEMICONDUCTOR INTRÍNSECO

−𝐸TUJ
𝜎 total = 𝜎j exp
2𝑘𝑇

El intervalo de energía prohibida depende del

parámetro de red del material
162
162

MOVILIDAD EN UN SEMICONDUCTOR
5000 1000
5000 4000 1000
4000
2000 NA # 1014
2000 NA # 1014
NA # 1016
"2$2
!n T
T "2$2 NA # 1016
1 1 1
!n 1000
2000 1000
2000 ND # 1014
ND # 1014
500 N # 1014 cm"3 = +
ND # 1016
ND # 1016
500 N #D1014 cm"3
100
100
D 200
200 T 500
500 1000
(K) 1000
NA # 1017
NA # 1017 𝜇 𝜇k 𝜇m
1000 T (K)
1000
!n (cm2/V-s)

!p (cm2/V-s)
17 N # 1018
!n (cm2/V-s)

!p (cm2/V-s)
N # 10 NA #A1018
ND D
# 1017
500 100
500 100

N # 1019
NA #A1019 1000
1000
18
N # 10
# 1018
ND D
500
500
T "2$2
Variación de la movilidad de electrones y
NDND
##1010
19
19 !p
!p T "2$2
huecos en silicio en función de la temperatura y
100
100
200 200 14
N # 10 cm
"3 en función de la concentración de dopantes. Los
NA # 10A 14 cm"3

100 100 ”insets” muestran la dependencia con la

100 100200 200 500 5001000 1000

5050 10 10
T (K)T (K)
temperatura para una muestra intrínseca.
"50
"50 00 5050 100
100 150150 200200 "50"50 0 0 50 50 100 100 150 150 200 200
T (!C)
T (!C) T )(!C)
T (!C
(a)(a) (b) (b)

Figure5.2
Figure 5.2| (a)
| (a)Electron
Electronand
and(b)(b)hole
hole mobilities
mobilities in in silicon
silicon versus
versus temperature
temperature forfor various
various doping
doping concentrations.
concentrations. Inserts
Inserts showshow

Dependencia de la movilidad con la temperatura:

temperaturedependence
temperature
(FromPierret
(From
dependenceforfor“almost”
Pierret[8].)
[8].)
“almost” intrinsic
intrinsic silicon.
silicon.

La movilidad de ve sólo afectada por la red cristalina: 𝜇k ∝ 𝑇 IC/B

n op/q
La movilidad de ve sólo afectada por impurezas ionizadas: 𝜇m ∝ , donde NI es el número de impurezas ionizadas
rs
 MOVILIDAD EN UN SEMICONDUCTOR 5.1 Carrier Drift 163

104
!n T ! 300 K

Ge
103 !p

102
104
Variación de la movilidad de electrones y
Mobility (cm2/V-s)

huecos en silicio y germanio en función de la

!n
3
concentración de impurezas.
10 Si
!p

102
104
!n

GaAs
103
!p

102
1014 1015 1016 1017 1018 1019
Impurity concentration (cm#3)

Figure 5.3 | Electron and hole mobilities versus impurity

concentrations for germanium, silicon, and gallium arsenide
at T ! 300 K.
(From Sze [14].)

If these two scattering processes are independent, then the total probability of a scatter-
 DEPENDENCIA DE LA RESISTIVIDAD CON LA TEMPERATURA EN UN SEMICONDUCTOR
5.1 Carrier Drift 167

T (K)
500
1000 300 200 100 75
1017
Electron concentration (cm"3)

1016 10
1 1

Conductivity (#– cm)"1

𝜌 = =
𝜎 𝑒 𝜇u 𝑛 + 𝜇J 𝑝
!
1015 1.0

n
Relación entre la concentración de electrones y
1014 0.1
la conductividad en función del inverso de la
ni temperatura en silicio.

1013
0 4 8 12 16 20
1000
(K"1 )
T

Figure 5.6 | Electron concentration and conductivity versus

inverse temperature for silicon.
(After Sze [14].)

We may plot the carrier concentration and conductivity of a semiconductor as

a function of temperature for a particular doping concentration. Figure 5.6 shows
the electron concentration and conductivity of silicon as a function of inverse tem-
perature for the case when Nd ! 1015 cm"3. In the midtemperature range, or extrinsic
range, as shown, we have complete ionization—the electron concentration remains
essentially constant. However, the mobility is a function of temperature so the con-
ductivity varies with temperature in this range. At higher temperatures, the intrinsic
carrier concentration increases and begins to dominate the electron concentration as